I got "when hell freezes over!"

If past experience is anything to go by, it will be much sooner.

---

A couple of MB splitters are back online.

---

I got "when hell freezes over!"

If past experience is anything to go by, it will be much sooner.

And with the summer (what summer) we've had up here north of the border my prediction may be sooner to!

Thx so much, all good stuff.

... and a big part of your complaint goes straight to the idea of "volunteer computing" -- volunteers aren't being paid, and may or may not care what happens to work they download.

Well, I'd never complain, as I'm just thrilled this project still exists (I was a member in 1999 as well), and that people everywhere can participate, and apologies I derailed the original topic of this thread as well.

You're spot on about various degrees of committment by participants (although it's the super committed who are tripping me up :-) ) and the unpredictability of reliable CPU labor (not limited to this project or even volunteerism), but there is also the aspect that some of the monster processors tend to carry a large carbon footprint upon us smaller guys --- how... because the nature of queueing when the deadlines of various WU's range from 3 days to 6 weeks with constant *backfilling* of yet more tasks means that outlying WU's upon download will not be priority until their due date is very very near. It's like if you set a deadline at a line in McDonalds to order before 2010, and keep letting people budge in front, you won't get your fries until 2010. As varying priorities are a fact, you should at least close ranks each time someone budges --- that way you fairly move up the queue.

So, my final thought on all this is (as I'm a small fish with a practical 3 WU queue limit) is I wish I could request WU's with deadlines within 3 days. But I know the nature of the chunker and empty queues (not enough work to satiate the masses -- what a great problem to have) doesn't afford that luxury.

Cheers

We're hoping a box of disks from AO will show up tomorrow and we can get some work generated.

---

@SETIEric@qoto.org (Mastodon)

Thanks for the update Eric. It is good to have an update.

---

We're hoping a box of disks from AO will show up tomorrow and we can get some work generated.

LOL...


SAAAAAY.....

How do we know that perfectly logical and simple explanation isn't just a smoke screen to cover for that pretty nice lookin' fish you're holdin' there?

You know what they say; "A bad day of fishin' beats a good day of system analysis (or even astronomy) every time!".

:-D

Alinator

How do we know that perfectly logical and simple explanation isn't just a smoke screen to cover for that pretty nice lookin' fish you're holdin' there?

Because no self-respecting fisherman would ever hide a magnificent fish that he'd caught with his own cunning and skill.

Smoke screen indeed! <ducking>

---

In two days I can't download wokunits for CPU and CUDA.
What can i do?

Thanks

Pray that UPS/FedEx/USPS brings the box of hard drives that was shipped from the telescope to Berkeley arrives in the lab tomorrow.

---

They ship the hard drives? Talk about high throughput on a slow data transfer! It's interesting to see that they ship hard drives back and forth, though I suppose satellite time to transmit them costs more.

---

They ship the hard drives? Talk about high throughput on a slow data transfer! It's interesting to see that they ship hard drives back and forth, though I suppose satellite time to transmit them costs more.

Much more expensive -- and it isn't fast enough.

Never underestimate the sheer amount of bandwidth available through FedEx.

---

LOL...

Yep, gives a whole new meaning to the term 'Burst Transmission'! :-D

Alinator

Something's happened.
Network traffic is maxed out, and i've gone from having about half a days work to almost a full cache again.

---

Grant
Darwin NT

Isn't a wokunit a little difficult to crunch? :-)

---

Something's happened.
Network traffic is maxed out, and i've gone from having about half a days work to almost a full cache again.

Yes, I'm confused & scared as well. :)

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[

Confused, scared, panicking.... :)

None of the above, but intrigued. For much of today, the only two tapes visibly 'in progress' have been our old friends 05mr09af and 27fe08ac - both of whom have moved on to channel two (genuinely - I got WUs from both tapes, and they seem to be crunching normally).

But those two tapes alone can't account for the 20/sec result creation rate, the maxxed-out line, and all the other WUs I've been getting (where's 15oc08ab, for example?).

All the work I got this morning came from a Sept and a Dec tape and they dont show at all on the splitter but I am sure someone will be able to inform me on the reason...my guess is an offline tape from storage?

---

http://www.novascotia.com

There also seems to be a problem with uploads AGAIN for the past few hours

There also seems to be a problem with uploads AGAIN for the past few hours

Could be related to the bandwidth being saturated with downloads? Hang in there - the storm will abate and give the uploads a chance eventually.

F.

---

Idle GPU cruncher..

I needed to delete ~ 30+ results, because the UL wasn't possible.*

[* it wasn't the first time that I needed to do this and will not be the last time..]

Then BOINC asked for new work..


Yes - either idle GPU cruncher or deleted results and again full loaded GPUs..

I'm curious how much WUs the GPU cruncher can catch before he will again idle because of too much ULs in BOINC..


Yes, it's not fair that at a GPU cruncher the '> CPUs x2' count for not work request..
It's not fair between CPU and GPU cruncher..

It would be better to make '> CUDA-cores x2' or something for GPU cruncher..

---